| { | |
| "best_metric": 0.7346938775510203, | |
| "best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-14/checkpoint-144", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 144, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 1.263960838317871, | |
| "learning_rate": 4.556569949810922e-06, | |
| "loss": 0.7021, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 1.1750551462173462, | |
| "learning_rate": 9.113139899621844e-06, | |
| "loss": 0.6857, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 1.3693294525146484, | |
| "learning_rate": 1.3669709849432766e-05, | |
| "loss": 0.6849, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.9440478086471558, | |
| "learning_rate": 1.822627979924369e-05, | |
| "loss": 0.6652, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.9551452398300171, | |
| "learning_rate": 2.2782849749054612e-05, | |
| "loss": 0.64, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.777751088142395, | |
| "learning_rate": 2.7339419698865533e-05, | |
| "loss": 0.6269, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 2.4816713333129883, | |
| "learning_rate": 3.1895989648676456e-05, | |
| "loss": 0.5586, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 1.696999430656433, | |
| "learning_rate": 3.645255959848738e-05, | |
| "loss": 0.6586, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 3.2336232662200928, | |
| "learning_rate": 4.10091295482983e-05, | |
| "loss": 0.6259, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_f1": 0.7326732673267327, | |
| "eval_loss": 0.7427406311035156, | |
| "eval_runtime": 1.4051, | |
| "eval_samples_per_second": 45.548, | |
| "eval_steps_per_second": 5.694, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 4.544165134429932, | |
| "learning_rate": 4.354055729819326e-05, | |
| "loss": 0.7, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 0.8700355887413025, | |
| "learning_rate": 4.3034271748214265e-05, | |
| "loss": 0.6972, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 1.7657314538955688, | |
| "learning_rate": 4.252798619823527e-05, | |
| "loss": 0.4085, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 0.6718289852142334, | |
| "learning_rate": 4.202170064825628e-05, | |
| "loss": 0.5618, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 1.5080432891845703, | |
| "learning_rate": 4.151541509827729e-05, | |
| "loss": 0.4822, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": Infinity, | |
| "learning_rate": 4.11103866582941e-05, | |
| "loss": 0.7679, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "grad_norm": 1.8398224115371704, | |
| "learning_rate": 4.0604101108315106e-05, | |
| "loss": 0.6183, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 1.7079250812530518, | |
| "learning_rate": 4.0097815558336114e-05, | |
| "loss": 0.5954, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 1.8899829387664795, | |
| "learning_rate": 3.959153000835712e-05, | |
| "loss": 0.596, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 3.904428243637085, | |
| "learning_rate": 3.908524445837813e-05, | |
| "loss": 0.6094, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_f1": 0.7326732673267327, | |
| "eval_loss": 0.673431396484375, | |
| "eval_runtime": 1.3605, | |
| "eval_samples_per_second": 47.041, | |
| "eval_steps_per_second": 5.88, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 1.0948293209075928, | |
| "learning_rate": 3.857895890839914e-05, | |
| "loss": 0.5749, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "grad_norm": 1.2664576768875122, | |
| "learning_rate": 3.807267335842015e-05, | |
| "loss": 0.4808, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "grad_norm": 1.7244511842727661, | |
| "learning_rate": 3.7566387808441155e-05, | |
| "loss": 0.5641, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 1.1561750173568726, | |
| "learning_rate": 3.706010225846216e-05, | |
| "loss": 0.5395, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": Infinity, | |
| "learning_rate": 3.665507381847897e-05, | |
| "loss": 0.5224, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 6.8102827072143555, | |
| "learning_rate": 3.614878826849998e-05, | |
| "loss": 0.4427, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "grad_norm": Infinity, | |
| "learning_rate": 3.574375982851679e-05, | |
| "loss": 0.6088, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "grad_norm": 3.278684139251709, | |
| "learning_rate": 3.52374742785378e-05, | |
| "loss": 0.3563, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "grad_norm": 4.377391815185547, | |
| "learning_rate": 3.4731188728558806e-05, | |
| "loss": 0.3177, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_f1": 0.7346938775510203, | |
| "eval_loss": 0.804656982421875, | |
| "eval_runtime": 1.3676, | |
| "eval_samples_per_second": 46.796, | |
| "eval_steps_per_second": 5.85, | |
| "step": 144 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 480, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 2121874430755872.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 4.374307151818485e-05, | |
| "per_device_train_batch_size": 4 | |
| } | |
| } | |